A apparatus for generating a depth map includes a cube space setting unit configured to set cube space based on camera intrinsic parameters; a layer generating unit configured to extract objects from a planar image and generate a layer including each object; a layer editing unit configured to apply at least one of parallel translation and rotation in a depth direction for the layer based on at least one of translation input and rotation input; and a model projecting unit configured to place a model on the layer, set a depth value for a model area placed on an object area included in the layer as a depth value of the object and generate a depth map comprising the depth value corresponding to each layer.